Luminescent materials are commonly used in security markings to be disposed on documents or articles, or in the bulk material of documents or articles, as an authenticity feature. A luminescent material typically converts energy of an exciting radiation of a given wavelength into emitted light having another wavelength. Luminescence emission used for authentication of a marking can lie in the spectrum range from UV light (200-400 nm), visible light (400-700 nm) or near to mid infrared light (700-2500 nm).
An “up-converter” material emits radiation at a shorter wavelength than that of the exciting radiation. By contrast, a “down-converter” material emits radiation at a longer wavelength than that of the exciting radiation. Most luminescent materials can be excited at more than one wavelength, and some luminescent materials can emit simultaneously at more than one wavelength.
Luminescence may be divided in: (i) phosphorescence, which relates to time-delayed radiation emission observable after the excitation radiation is removed (typically, with a decay lifetime from above about 1 μs to about 100 s), and (ii) fluorescence, which relates to prompt radiation emission upon excitation (typically, with a decay lifetime below 1 μs).
Thus, a luminescent material, upon illumination with excitation light within a first wavelength range, emits luminescence light within a second wavelength range, which may differ from or overlap with said first wavelength range, depending on the luminescent material used. A characteristic spectral property of a luminescent material such as its emission light intensity profile with time, or its characteristic decay time after excitation has stopped, for example, constitutes a signature of a presence of this very material and may thus be used as an authenticity feature for detecting genuineness or forgery of a luminescent marking.
Luminescent materials are classic ingredients of security inks or coatings. For example, the following patents disclose luminescent substances, which may include mixtures of pigments having distinct decay time properties, and security paper including such substances: EP 0 066 854 B1, U.S. Pat. No. 4,451,530, U.S. Pat. No. 4,452,843, U.S. Pat. No. 4,451,521. Processes and apparatuses for detecting luminescence light and authenticity of a marked item are also well known: see, for example, U.S. Pat. No. 4,598,205, or U.S. Pat. No. 4,533,244, which disclose sensing decay behavior of luminescence emissions. Luminescent coded symbols are known from U.S. Pat. No. 3,473,027, and an optical reader for luminescent codes is disclosed in U.S. Pat. No. 3,663,813. The patents U.S. Pat. No. 6,996,252 B2, U.S. Pat. No. 7,213,757 B2 and U.S. Pat. No. 7,427,030 B2 disclose using two luminescent materials, having distinct decay time properties, for authenticating an item.
Typically, a scanner according to the known art for detecting time-dependent luminescence light comprises a power source, or a connection to a power source, a light source (connected to the power source) for illuminating a luminescent material with excitation light, a light sensor for measuring an intensity of the luminescence light emitted by the luminescent material, and a control unit (processor) for controlling the power source, light source and light sensor to acquire and store an intensity profile of the emitted luminescence light, and calculating a decay time value from this intensity profile. A luminescence emission intensity profile (or intensity-versus-time curve) comprises successive emission intensity values I(t1), . . . , I(tn) from a luminescent material measured at the consecutive times t1, . . . , tn, together forming a luminescence emission curve I(t).
The light source in such scanner, depending on the part of the spectrum used for the detection of the luminescent material, may be an incandescent lamp, typically for wavelengths between about 400 nm to about 2500 nm, used with mechanical or opto-electronic devices for delivering pulsed light, or a flash lamp (e.g. a Xenon high-pressure flash lamp), or a laser or Light-Emitting-Diode (LED), emitting in the UV, visible or IR region, typically for wavelengths from about 250 nm to about 1 μm. The light source may be powered by a drive current (for a LED for example) or by a drive voltage (for a discharge lamp, for example).
The light sensors or photodetectors in such scanner may be photodiodes (single or arrays), phototransistor or photoresistance circuits, linear CMOS or CCD sensors, for example.
The scanner, in addition to its specific power module for supplying the scanner with power, may also comprise a communication module, which may be a radio module for wireless communication (over Wi-Fi, for example), a display module, e.g. a liquid crystal display LCD, or kinescope display, for displaying measured data or scanning parameters, and a controlling interface for inputting scan conditions, including control switches having multiple functions and an ON/OFF switch.
Most often, the time dependent intensity curve of luminescence emission light (i.e. intensity profile with time) from a luminescent material is modeled by an exponential law I(t)=I0 exp(−α[t−t0]), wherein time t is counted from initial time t0 at which the excitation light illuminating said material is switched off. Thus, obtaining a value corresponding to the decay rate constant α characterizing the luminescent material necessitates measuring, after excitation has stopped, an emission intensity profile composed of successive emission intensity values I(t1), . . . , I(tn), over a time interval. The decay time value τ from the intensity profile I(t) corresponds to α−1. In commercially available scanners, a pulsed light source illuminates the luminescent material with an excitation light of a given intensity, in a first wavelength range, during an excitation time interval. After the illumination has stopped, possibly with a time delay, the light sensor starts measuring successive values of the decaying luminescence light intensity in a second wavelength range over a measuring time interval, and the corresponding luminescence intensity profile is stored in a memory. The operation may be repeated so as to obtain more reliable average values. Usually, it is possible to set the excitation time interval and/or the time delay so as to avoid measuring values of luminescence intensity below a detection threshold value of the light sensor or above its saturation threshold value.
However, variants are also known. For example, U.S. Pat. No. 6,264,107 B1 discloses determining a decay time from the time required for the latent phosphorescence intensity to fall through two predetermined thresholds. This patent discloses a scanner comprising a flood LED (FLED) as a light source, i.e. a very intense light source. Such an intense light source is indeed necessary in this case for charging enough a tag comprising the luminescent material (phosphor) and preventing the problem of low signal response.
In an other approach, U.S. Pat. No. 7,262,420 B1 discloses carrying out multiple illuminations with excitation light for obtaining a single decay time value: the light source is successively activated (during a same excitation time interval) and a single measure of luminescence intensity is performed after the illumination of the luminescent material with the excitation light source has been switched off, but each successive measurement is performed with a different time delay counted from the time at which excitation light is switched off. However, this method is time consuming, as it requires one illumination per measured intensity value. Moreover, in order to obtain more reliable results, this method requires repeated measurements corresponding to a same time delay.
In order to obtain a stronger luminescence signal, some scanners allow setting the excitation time interval, so as to “charge” enough the luminescent particles in the luminescent material (i.e. excite a great number of such luminescent particles to obtain more intense luminescence emission). Moreover, for a better accuracy of the determined decay time value, a plurality of valid intensity profiles are successively acquired (for example, about a hundred), these curves are then summed and an average curve is calculated. Increased accuracy is obtained if the resulting signal of the measured intensity is normalized and the normalized signal is used for calculating the decay time value. An intensity profile is valid if the intensity value of at least the first point of the intensity profile is above a detection threshold of the light sensor and below its saturation threshold (if said value is too low or too high, the excitation time is respectively increased or decreased). However, a problem arises in case where the excitation time interval is too short for allowing a reliable normalization of the luminescence intensity signal, particularly for luminescent materials including a mixture of luminescent particles of different types and of which decay time values are widely differing (for example, particles having the shortest decay time may not be detected by the scanner). Another problem arising with the use of variable excitation time is that the luminescent material is not excited under the same conditions for all the intensity profiles, and, in case of a material including a mixture of luminescent pigments having distinct decay time properties, this may cause confusion. For example, FIGS. 1A and 1B illustrate a case of normalized intensity profiles from a marking with an ink (luminescent material) including two types of luminescent pigments: pigments P1 and P2; in this example, the decay time value of pigments P1 is about 100 μs, and the decay time value of pigments P2 is about 500 μs. FIG. 1A shows an excitation curve (1) having a long excitation time interval of 100 μs, and a corresponding normalized luminescence intensity profile (P1+P2) for a luminescent ink material being a mixture of 50% of the first pigments P1 (having the shorter decay time, or higher decay rate) and 50% of the second pigments P2 (having the longer decay time, or lower decay rate).
FIG. 1B corresponds to a luminescent ink including a mixture of 42% of the first pigments P1 and 58% of the second pigments P2. In this case, the excitation time has been adjusted to a shorter value of 10 μs, as shown on excitation curve (2). Although the concentrations of pigments in the mixtures of pigments P1 and P2 significantly differ from FIG. 1A to FIG. 1B, the normalized luminescence intensity profiles (P1+P2) are very similar and can hardly be distinguished. Thus, it is not always possible, or it may be difficult, to detect a difference between two mixtures on the basis of luminescence intensity profiles obtained by varying the excitation time. Although the above example relates to pigments having typical decay time values of about few hundreds of microseconds, a similar conclusion remains for pigments having much longer decay time values (few ms or more).
Another problem arising with said known decay time scanners is that they do not allow acquiring a luminescence intensity profile, and thus determining a corresponding decay time value, or decay time values and also concentrations in case of a mixture of different types of pigments, in case the luminescent material is moving past the scanner; particularly, in case the luminescent material moves fast past the scanner on a production/distribution line. For example, in case of items marked with a luminescent material and transported on a conveyor belt of a production line moving with typical speed of about 200 to about 400 m/min (i.e. about 3 to 6 m/s), it is clearly not possible to acquire a luminescent intensity profile I(t), even is the luminescent material has quite a long decay time value of a few ms or more. Thus, identification/authentication of said moving marking is not possible in-line: for example, authenticating a luminescent marking such as a barcode or a datamatrix on an item transported on a conveyor belt. Consequently, in-line secure track and trace operations based on such in-line determination of a luminescent intensity profile are not possible, although highly desirable.